Research On Resource Optimization Of Heterogeneous Cellular Network

With the rapid development of Internet of Things applications such as In-dustry 4.0 and Internet of Vehicles,traditional cellular networks are difficult to meet the needs of forthcoming massive,diverse and green network connections.A large number of micro cells and relay nodes are deployed in the traditional macrocells to form a heterogeneous cellular multi-layer network,thus the en-ergy consumption of the base station and the terminal is reduced by shortening the data transmission link and the capacity and spectrum efficiency of the sys-tem are improved through space multiplexing.However,the number of base stations and the deployment cost of the network are increased,which leads to a sharp increase in energy consumption and the operating cost of operators' elec-tricity bills.Furthermore,in order to meet the high-speed data requirements of indoor IoT users,millimeter-wave massive multiple input multiple output system is adopted in heterogeneous micro cells to achieve parallel data trans-mission.However,massive antenna array needs a large number of RF links to drive,which increases the hardware design complexity,cost and energy con-sumption.In this paper,we optimize the capacity,energy consumption and cost of heterogeneous cellular networks by means of resource allocation.Through us-ing graph theoiy,auction theory and multi-objective genetic algorithm,wireless resource allocation of uplink and downlink,relay deployment and transmission scheduling,millimeter wave large-scale multi-antenna transmission technology and energy management of base station are researched,then a series of solutions and algorithms are derived.The main work and innovations of this paper are as follows:1?In order to solve the problem of capacity and energy consumption in two-tier heterogeneous cellular networks,resource allocation strategies for up-link and downlink transmission are proposed based on multi-objective opti-mization method in the inter-frequency networking mode.In the downlink transmission,the dynamic spectrum sharing strategy based on auction theory is designed and the spectrum-for-energy compensation incentive mechanism is used to design the utility function.Based on the NSGA-? multi-objective optimization algorithm,the utility of the macrocell and the femtocell is bal-anced.For the data uplink transmission,a multi-objective optimization algo-rithm based on the NSGA-? and the MOMA is designed to realize the alloca-tion of the joint user,the channel and the power to balance the energy efficiency and the spectral efficiency of the system.The experimental results show that the proposed dynamic spectrum sharing strategy and multi-objective optimized resource allocation scheme can balance the utility between heterogeneous net-works and the overall spectrum efficiency and energy efficiency of the system respectively.2.In order to solve the deployment cost optimization problem of heteroge-neous relay cellular network and the capacity optimization of hybrid data trans-mission,we propose the optimal deployment algorithm considering green relay location and number simultaneously and coordinated data transmission schedul-ing optimization strategy for heterogeneous cellular network respectively.A distributed 1.61 approximation algorithm based on dual planning method is de-signed to minimize network deployment cost and V2R/V2V collaborative data transmission scheduling and channel allocation strategy based on graph theory and semi-definite programming method is proposed.The experimental results show that the distributed 1.61 approximate optimization algorithm is superior in terms of complexity and total deployment cost,while the cooperative data transmission scheduling algorithm makes full use of the multi-channel charac-teristics of the heterogeneous network,and the service capacity is better than the IF algorithm based on data broadcast strategy and the CDD algorithm based on data transmission scheduling in single channel environment.3?To jointly optimize hybrid precoding of transmitter and receiver at a given user rate requirement in massive rmillimeter-wave single-user MIMO sys-tem of microcell,a hybrid precoding strategy based on singular value decompo-sition(SVD)of antenna array response matrix and multi-obj ective optimization based on sparse reconstruction are proposed.Firstly,a hybrid precoding strat-egy with variable number of RF links is designed.The number of RF links at the transmitter and receiver are selected according to the rate requirement,and the base vector with the largest correlation with the all-digital precoding matrix is choosen from the right singular vector matrix to form the hybrid precoding matrix.Moreover,the NSGA-? and ALSN multi-objective optimization al-gorithms are used to jointly optimize the number of RF links and the hybrid precoding matrix to balance the energy efficiency and spectrum efficiency of the system.The experimental results show that the algorithm based on singular value decomposition(SVD)can avoide the complex iterative loop search pro-cess in OMP algorithm.And the algorithm based on multi-objective optimiza-tion can avoide traversing all combinations of link numbers to find a strategy for optimal link number combination,which is better than the OMP algorithm under the same number of RF links.4?In order to deal with the problem of rapid increasement of energy con-sumption and power expenditures due to large-scale deployment of base sta-tions in heterogeneous cellular networks,an online energy scheduling strategy is proposed in the scenario of hybrid energy powered base stations including renewable energy and traditional grid.Firstly,the base station model and the queue model are constructed,and then the objective function of queue stabil-ity and power expenditure equilibrium is constructed by using the minimum condition Lyapunov drift and weighted penalty function,and finally it can be converted into linear programming problem to solve.The experimental results show that the method can obtain online energy scheduling optimization with-out predicting the arrival of data tasks,renewable energy output and real-time electricity price,respectively.Compared with the case of no renewable en-ergy and without scheduling strategy,it can decrease by 68.63%and 31.64%,respectively.